Metalab researcher, Zheng Wang, has won the best paper award in The International Symposium on Code Generation and Optimization (CGO) 2017.

The paper is co-authored by Zheng with Chris Cummins, Pavlos Petoumenos and Hugh Leather from University of Edinburgh.

This work develops a deep learning based system, CLGen, to mine code from GitHub to automatically generate benchmark programs that are indistinguishable from human code. This provides a way to automatically generate an unbounded number of benchmarks for testing and designing computing systems.

CLGen is opensourced and can be downloaded from

CGO is a top tier conference in compilation and code optimisation. “It provides a premier venue to bring together researchers and practitioners working at the interface of hardware and software on a wide range of optimization and code generation techniques and related issues. The conference spans the spectrum from purely static to fully dynamic approaches, and from pure software-based methods to specific architectural features and support for code generation and optimization.”

CGO 2017 accepts 26 papers out of 114 submission with an acceptance rate of 22%.